Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A system, comprising: an image sensor disposed about an autonomous vehicle and configured to sense an environment and capture image data; a LIDAR sensor configured to capture LIDAR data; and one or more processors communicatively coupled to the image sensor and the LIDAR sensor, the one or more processors being configured to perform operations comprising: receiving a signal from the image sensor, the signal comprising data representing a sensor measurement of the environment; determining, based at least in part on the data, that the image sensor is miscalibrated, wherein determining that the image sensor is miscalibrated comprises: determining, based at least in part on the LIDAR data, a set of detected edges associated with an object in the environment; and determining region data associated with the set of detected edges; generating an expected sensor measurement, the expected sensor measurement associated with the image sensor; generating a calibration parameter, associated with the image sensor, based at least in part on the expected sensor measurement; and modifying a parameter of the image sensor based at least in part on the calibration parameter and a parameter associated with the region data.
The system relates to autonomous vehicle calibration, specifically addressing miscalibration in image sensors used for environmental perception. Autonomous vehicles rely on accurate sensor data for navigation and obstacle detection, but image sensors can become miscalibrated due to environmental factors or mechanical shifts, leading to inaccurate measurements. The system includes an image sensor and a LIDAR sensor mounted on the vehicle to capture environmental data. The system processes this data using one or more processors to detect miscalibration in the image sensor. The processors analyze LIDAR data to identify edges of objects in the environment and generate region data associated with these edges. The system then compares the image sensor's measurements with an expected sensor measurement derived from the LIDAR data. If a discrepancy is found, a calibration parameter is generated to adjust the image sensor's parameters, ensuring accurate alignment with the LIDAR-derived region data. This calibration process improves the reliability of the vehicle's perception system by dynamically correcting misalignments between the image sensor and other sensors. The system enhances autonomous vehicle safety and performance by maintaining precise sensor calibration in real-time.
2. The system of claim 1 , wherein the LIDAR sensor is a first LIDAR sensor, wherein the system further comprises a plurality of additional LIDAR sensors, and the operations further comprising determining that the first LIDAR sensor is miscalibrated, and wherein determining that the first LIDAR sensor is miscalibrated comprises: determining a first sensor measurement of the first LIDAR sensor comprising a first distance or a first reflectivity value; determining a plurality of additional sensor measurements of the plurality of additional LIDAR sensors comprising a second distance or a second reflectivity value, the first sensor measurement and the plurality of additional sensor measurements being associated with an object in the environment; and determining a difference between the first sensor measurement and at least one measurement of the plurality of additional sensor measurements, and the operations further comprising modifying a parameter of the first LIDAR sensor, based at least in part on the difference.
The system involves a network of LIDAR sensors used for environmental perception, such as in autonomous vehicles or robotics, where accurate calibration is critical for reliable object detection and distance measurement. The problem addressed is the potential miscalibration of individual LIDAR sensors, which can lead to inaccurate data and system failures. The system includes a primary LIDAR sensor and multiple additional LIDAR sensors. To detect miscalibration, the system compares measurements from the primary sensor (e.g., distance or reflectivity values) with corresponding measurements from the additional sensors for the same object in the environment. If a significant difference is detected between the primary sensor's measurement and those of the additional sensors, the system determines that the primary LIDAR sensor is miscalibrated. Upon detecting miscalibration, the system adjusts a parameter of the primary LIDAR sensor to correct the discrepancy, ensuring consistent and reliable data across the sensor network. This self-calibration mechanism improves system accuracy and reduces the need for manual recalibration.
3. The system of claim 1 , wherein modifying the parameter of the image sensor comprises adjusting a focus parameter or a lens parameter of the image sensor.
This invention relates to image capture systems, specifically addressing the challenge of optimizing image quality by dynamically adjusting image sensor parameters. The system includes an image sensor configured to capture images and a processor that modifies at least one parameter of the image sensor to enhance image quality. The modification involves adjusting either a focus parameter or a lens parameter of the image sensor. The focus parameter may include settings such as autofocus speed, focus range, or focus accuracy, while the lens parameter may include aperture size, focal length, or lens distortion correction. The processor may analyze captured images or sensor data to determine optimal adjustments, ensuring improved clarity, sharpness, or overall image performance. This dynamic adjustment capability allows the system to adapt to varying lighting conditions, subject movement, or environmental factors, resulting in higher-quality images. The invention is particularly useful in applications requiring real-time image optimization, such as surveillance, automotive imaging, or consumer electronics.
4. The system of claim 1 , wherein the system further comprises an IMU configured to capture IMU sensor data and a GPS configured to capture GPS sensor data, and wherein generating the expected sensor measurement comprises: fusing at least two of the image data, the LIDAR data, the IMU sensor data, or the GPS sensor data to generate fused data; receiving, based at least in part on the fused data, map data, the map data indicative of an object in the environment having known properties; and generating the expected sensor measurement based at least in part on the object.
A system for environmental perception in autonomous vehicles or robotic applications integrates multiple sensors to improve accuracy in detecting and tracking objects. The system combines image data from cameras, LIDAR data from laser scanners, inertial measurement unit (IMU) sensor data, and global positioning system (GPS) sensor data. These inputs are fused into a unified dataset to enhance spatial and temporal coherence. The fused data is then compared against map data, which contains known properties of objects in the environment, such as their location, shape, and reflectivity. By leveraging this map data, the system generates an expected sensor measurement for the object, allowing for more reliable detection and tracking. The IMU provides motion and orientation data, while the GPS offers positional information, both of which contribute to refining the system's understanding of the environment. This approach reduces errors caused by individual sensor limitations, such as occlusions in camera images or noise in LIDAR readings, by cross-referencing multiple data sources. The system is particularly useful in dynamic environments where real-time object recognition and localization are critical for navigation and decision-making.
5. The system of claim 1 , wherein generating the calibration parameter comprises: receiving additional data from a log file, a map, or an additional sensor disposed about the system; and generating, based at least in part on the additional data, the calibration parameter, wherein the calibration parameter is configured to alter an intrinsic value associated with the image sensor.
This invention relates to a calibration system for an image sensor, addressing the challenge of maintaining accurate image data by dynamically adjusting sensor parameters based on external inputs. The system receives additional data from various sources, including log files, maps, or supplementary sensors integrated into the system. This data is used to generate a calibration parameter that modifies an intrinsic value of the image sensor, such as exposure time, gain, or pixel sensitivity. The calibration process ensures the sensor adapts to environmental or operational changes, improving image quality and reliability. The system may also incorporate data from multiple sources to refine the calibration parameter, enhancing precision. This approach is particularly useful in applications where environmental conditions or sensor performance drift over time, such as autonomous vehicles, surveillance systems, or industrial imaging. By dynamically adjusting sensor parameters, the system mitigates errors and ensures consistent image output. The calibration parameter can be applied in real-time or during post-processing, depending on system requirements. The invention improves upon traditional static calibration methods by incorporating real-time or near-real-time adjustments based on external data, leading to more accurate and adaptable image capture.
6. The system of claim 5 , wherein the additional data comprises an image captured by a camera, a velocity of the system captured by an odometry sensor, a position captured by a global positioning (GPS) sensor, audio data captured by a microphone, sound navigation and ranging (SONAR) data captured by a SONAR sensor, radio detection and ranging (RADAR) data captured by a RADAR sensor, direction of gravity data captured by an inertial measurement unit (IMU), or a direction of travel data.
This invention relates to a system for collecting and processing sensor data to enhance situational awareness, navigation, or environmental mapping. The system integrates multiple sensors to gather diverse data types, including visual, auditory, and spatial information, to improve accuracy and reliability in applications such as autonomous navigation, robotics, or environmental monitoring. The system captures images using a camera, velocity data from an odometry sensor, and positional data from a GPS sensor. It also records audio data via a microphone and spatial data using SONAR and RADAR sensors. Additionally, the system collects directional data from an inertial measurement unit (IMU) and determines the direction of travel. These data types are combined to provide a comprehensive understanding of the environment, enabling real-time decision-making or mapping. The integration of these sensors allows the system to compensate for individual sensor limitations, such as GPS signal loss or camera occlusion, by cross-referencing multiple data sources. This redundancy improves accuracy and robustness, particularly in dynamic or challenging environments. The system may be used in autonomous vehicles, drones, or robotic systems where precise environmental awareness is critical. The collected data can be processed to generate maps, detect obstacles, or navigate autonomously, enhancing safety and efficiency in various applications.
7. The system of claim 1 , wherein the operations further comprise: generating, based at least in part on the calibration parameter, a vehicle trajectory, the vehicle trajectory configured to cause the system to traverse a portion of the environment.
This invention relates to autonomous vehicle navigation systems designed to improve path planning and obstacle avoidance in dynamic environments. The system addresses challenges in accurately navigating complex environments by dynamically adjusting vehicle trajectories based on real-time calibration parameters. These parameters are derived from sensor data, environmental conditions, and vehicle dynamics to optimize path planning. The system generates a vehicle trajectory that guides the vehicle through a portion of the environment while avoiding obstacles and adhering to operational constraints. The trajectory is continuously updated to ensure safe and efficient navigation. The system integrates sensor inputs, such as LiDAR, cameras, and radar, to refine calibration parameters and adjust the trajectory in response to changing conditions. This approach enhances the vehicle's ability to navigate unpredictable environments, reducing the risk of collisions and improving overall navigation efficiency. The invention is particularly useful in autonomous driving, robotics, and unmanned aerial vehicles where precise path planning is critical.
8. The system of claim 1 , wherein the operations further comprise: generating a first probabilistic model; and generating, based at least in part on a heuristic rule, a second probabilistic model, wherein generating the calibration parameter comprises using the first probabilistic model, and wherein modifying the image sensor comprises using the second probabilistic model.
This invention relates to image processing systems that use probabilistic models to calibrate and modify image sensors. The system addresses the challenge of accurately calibrating image sensors to improve image quality by leveraging both data-driven and rule-based approaches. The system generates a first probabilistic model, which is used to determine calibration parameters for the image sensor. Additionally, a second probabilistic model is generated based on heuristic rules, which is then used to modify the image sensor. The combination of these models allows for more precise and adaptive calibration, improving the sensor's performance under varying conditions. The system may involve training the probabilistic models using historical sensor data or predefined rules, ensuring that the calibration and modifications are both statistically robust and aligned with domain-specific knowledge. This dual-model approach enhances the accuracy and reliability of image sensor adjustments, particularly in applications requiring high precision, such as medical imaging, autonomous vehicles, or industrial inspection. The invention aims to optimize sensor performance by integrating probabilistic learning with heuristic-based refinements, reducing errors and improving consistency in image capture.
9. A method, comprising: receiving a plurality of sensor measurements from a plurality of sensors disposed about a vehicle, wherein the plurality of sensors comprise a LIDAR sensor configured to capture LIDAR sensor data and an image sensor configured to capture image sensor data; identifying, based at least in part on the plurality of sensor measurements, an anomalous sensor measurement of the plurality of sensor measurements associated with the image sensor of the plurality of sensors, wherein identifying the anomalous sensor measurement comprises: determining a set of edges in the LIDAR sensor data; associating, as region data, at least a portion of the image sensor data with the LIDAR sensor data; and identifying a portion of the region data that is anomalous; determining, based at least in part on the anomalous sensor measurement, that the image sensor is miscalibrated; generating an expected sensor measurement, associated with the image sensor; and generating, based at least in part on the expected sensor measurement, a calibration parameter for calibrating the image sensor.
This invention relates to vehicle sensor calibration, specifically addressing miscalibration in image sensors by leveraging LIDAR data. The problem solved is the detection and correction of miscalibrated image sensors in vehicles equipped with multiple sensors, including LIDAR and cameras. The method involves receiving sensor measurements from various sensors, including LIDAR and image sensors. It identifies anomalous measurements from the image sensor by comparing image data with LIDAR data. This comparison involves detecting edges in LIDAR data, associating corresponding image data with the LIDAR data, and identifying discrepancies in the image data that indicate miscalibration. Once an anomalous measurement is detected, the system determines that the image sensor is miscalibrated. It then generates an expected sensor measurement for the image sensor and uses this expected measurement to derive a calibration parameter. This parameter is applied to correct the image sensor's calibration, ensuring accurate sensor data for vehicle systems. The approach improves sensor reliability by cross-referencing LIDAR and image data to detect and rectify calibration errors.
10. The method of claim 9 , further comprising: generating fused sensor data using data received from the LIDAR sensor and the image sensor, the fused sensor data indicative of a localization of the vehicle; and generating a probabilistic map of an environment based at least in part on the fused sensor data, the probabilistic map including a plurality of labeled objects and a probability score assigned to at least one object of the plurality of labeled objects, the plurality of labeled objects being associated with object parameters, wherein identifying the anomalous sensor measurement comprises comparing data received from the image sensor or the LIDAR sensor with the object parameters.
This invention relates to autonomous vehicle navigation systems that use sensor fusion and probabilistic mapping to improve localization and anomaly detection. The system integrates data from a LIDAR sensor and an image sensor to generate fused sensor data, which provides the vehicle's position and orientation. A probabilistic map of the environment is then created using this fused data, containing labeled objects with associated parameters and probability scores indicating confidence in their detection. The system detects anomalous sensor measurements by comparing real-time sensor data against the expected object parameters from the probabilistic map. This approach enhances reliability by cross-referencing multiple sensor inputs and leveraging probabilistic modeling to identify discrepancies. The method ensures accurate localization and object recognition while flagging potential sensor errors or environmental changes that could affect autonomous driving decisions. The probabilistic scoring allows the system to adapt to varying confidence levels in object detection, improving robustness in dynamic environments. This technique is particularly useful for autonomous vehicles operating in complex or uncertain conditions where sensor accuracy may be compromised.
11. The method of claim 10 , wherein generating the probabilistic map comprises: accessing historical sensor data corresponding to the plurality of sensors; and generating the probability score based at least in part on the historical sensor data and the anomalous sensor measurement.
This invention relates to a method for generating a probabilistic map to identify anomalies in sensor data. The method addresses the problem of detecting and evaluating anomalies in sensor measurements, which is critical for applications such as industrial monitoring, predictive maintenance, and environmental sensing. The probabilistic map helps assess the likelihood of anomalies by incorporating both real-time sensor data and historical sensor data. The method involves accessing historical sensor data from a plurality of sensors and generating a probability score based on this historical data along with an anomalous sensor measurement. The historical sensor data provides context for evaluating the significance of the anomaly, allowing for more accurate probabilistic assessments. By comparing current anomalous measurements against historical patterns, the method improves anomaly detection accuracy and reduces false positives. The probabilistic map is generated by analyzing the relationship between the anomalous measurement and the historical data, assigning a probability score that reflects the likelihood of the anomaly being significant. This approach enhances decision-making in systems where sensor reliability and anomaly detection are critical, such as in industrial automation, smart infrastructure, and IoT networks. The method ensures that anomalies are evaluated in context, improving the overall robustness of sensor-based monitoring systems.
12. The method of claim 9 , wherein identifying the portion of the region data that is anomalous comprises identifying a portion of the region data that is blurry.
This invention relates to image processing, specifically detecting anomalies in region data, such as identifying blurry portions within an image. The problem addressed is the need for automated systems to accurately detect and isolate regions of poor image quality, such as blurriness, which can affect subsequent analysis or user experience. The method involves analyzing region data, which may be derived from an image or other visual input, to determine if any portion exhibits characteristics of blurriness. This is achieved by evaluating the region data for visual degradation, such as lack of sharpness or clarity, which distinguishes it from normal, high-quality regions. The detection process may involve comparing the region data against predefined quality thresholds or using computational techniques to assess blur metrics. Once identified, the blurry portion can be flagged for further processing, such as correction, exclusion, or user notification. The method may also include preprocessing steps to enhance the accuracy of blur detection, such as noise reduction or contrast adjustment. Additionally, the system may apply adaptive thresholds or machine learning models trained to recognize various types of blurriness, ensuring robust performance across different image types and conditions. The output of this process is a clear indication of which regions are anomalous due to blurriness, enabling downstream applications to handle them appropriately. This improves the reliability of image-based systems in fields such as medical imaging, surveillance, and autonomous navigation.
13. The method of claim 9 , wherein the LIDAR sensor is configured to capture a first sensor measurement and an additional LIDAR sensor is configured to capture a second sensor measurement, the first sensor measurement and the second sensor measurement associated with a distance of an object to the vehicle or a reflectivity of the object, and wherein identifying the anomalous sensor measurement comprises comparing the first sensor measurement and the second sensor measurement.
This invention relates to a system for detecting anomalous sensor measurements in autonomous vehicle navigation, particularly using LIDAR sensors. The problem addressed is ensuring accurate object detection and distance measurement by identifying discrepancies between multiple LIDAR sensors. The system employs at least two LIDAR sensors to capture measurements of an object's distance or reflectivity relative to the vehicle. By comparing the first sensor's measurement with the second sensor's measurement, the system identifies any anomalous readings that deviate significantly from expected values. This comparison helps filter out erroneous data caused by environmental factors, sensor malfunctions, or other interference, improving the reliability of autonomous vehicle navigation. The method involves analyzing the consistency between the two measurements to determine if one or both readings are unreliable, thereby enhancing the accuracy of object detection and distance assessment in real-time driving scenarios. The approach ensures that the vehicle's perception system operates with higher fidelity by cross-referencing multiple sensor inputs.
14. The method of claim 9 , wherein the vehicle is an autonomous vehicle, the method further comprising: generating, based on the calibration parameter, a trajectory configured to cause the autonomous vehicle to traverse an environment, and transmitting the trajectory to the autonomous vehicle.
This invention relates to autonomous vehicle calibration and trajectory generation. The technology addresses the challenge of ensuring accurate sensor calibration in autonomous vehicles to enable precise navigation and safe operation in dynamic environments. The method involves calibrating a sensor system on an autonomous vehicle by determining a calibration parameter that corrects misalignment or other errors in sensor data. This calibration parameter is then used to generate a trajectory that guides the autonomous vehicle through an environment, ensuring that the vehicle follows a path that accounts for the calibrated sensor data. The trajectory is transmitted to the autonomous vehicle, allowing it to navigate with improved accuracy and reliability. The calibration process may involve analyzing sensor data, comparing it to reference data, and adjusting the calibration parameter iteratively until the sensor system meets performance criteria. The generated trajectory ensures the vehicle avoids obstacles, adheres to road conditions, and maintains safe operation. This approach enhances the autonomy and safety of self-driving vehicles by integrating real-time calibration with trajectory planning.
15. One or more non-transitory computer readable media having instructions stored thereon which, when executed by one or more processors of a system, cause the system to perform operations comprising: receiving first sensor data associated with a first measurement made by a first sensor disposed about the system, the first sensor comprising an image sensor; receiving second sensor data associated with a second sensor disposed about the system, the second sensor comprising a LIDAR sensor; determining, based at least in part on the first sensor data and the second sensor data, that the first sensor data is anomalous, wherein determining that the first sensor data is anomalous comprises: detecting a set of detected edges in the second data; determining, as region data, at least a portion of the first sensor data with the set of detected edges; and determining, based at least in part on the region data, that the first sensor data is anomalous; and determining an expected sensor measurement based at least in part on the first measurement and the second sensor data; determining a calibration parameter associated with the first sensor based at least in part on the expected sensor measurement; and calibrating the first sensor based at least in part on the calibration parameter.
The invention relates to a system for calibrating an image sensor using data from a LIDAR sensor to detect and correct anomalies in the image sensor's measurements. The system addresses the problem of ensuring accurate sensor data in environments where image sensors may produce unreliable or anomalous readings, such as due to environmental conditions or sensor degradation. The system receives image data from an image sensor and LIDAR data from a LIDAR sensor. It analyzes the LIDAR data to detect edges in the environment, then uses these edges to identify corresponding regions in the image data. By comparing the image data within these regions to expected values derived from the LIDAR data, the system determines if the image data is anomalous. If anomalies are detected, the system calculates an expected sensor measurement by combining the image data and LIDAR data. It then derives a calibration parameter to adjust the image sensor's output, ensuring more accurate future measurements. This approach improves sensor reliability by cross-referencing data from multiple sensor types, reducing errors in image-based measurements.
16. The one or more non-transitory computer readable media of claim 15 , wherein determining that the first sensor data is anomalous comprises determining that the first sensor data is blurry.
A system for analyzing sensor data in industrial or environmental monitoring applications detects anomalies in sensor readings to improve reliability and accuracy. The system processes sensor data from multiple sources, such as cameras or environmental sensors, to identify deviations from expected patterns. Specifically, the system determines whether sensor data is anomalous by assessing whether the data is blurry, which may indicate poor image quality, sensor malfunction, or environmental interference. The system compares the sensor data against predefined thresholds or machine learning models trained to recognize normal and abnormal data patterns. When blurriness is detected, the system may trigger corrective actions, such as recalibrating the sensor, alerting operators, or discarding unreliable data. The system may also integrate data from multiple sensors to cross-validate findings and reduce false positives. This approach enhances the robustness of sensor networks in applications like manufacturing, surveillance, or environmental monitoring, where accurate and reliable data is critical for decision-making. The system may be implemented as part of a larger monitoring framework, ensuring continuous and automated anomaly detection.
17. The one or more non-transitory computer readable media of claim 15 , wherein the second data comprises second sensor data acquired from a plurality of additional sensors comprising an additional LIDAR sensor, an additional image sensor, a GPS sensor, an IMU sensor, or a rotary encoder, the operations further comprising: generating fused sensor data using the first sensor data and the second sensor data, the fused sensor data indicative of a localization of the system; and accessing, based at least in part on the localization, map data, the map data indicative of an object in an environment, the object associated with one or more object properties, wherein determining that the first sensor data is anomalous comprises determining that the first sensor data differs from the one or more object properties.
This invention relates to sensor data fusion and anomaly detection in autonomous systems, particularly for improving localization and environmental awareness. The system acquires first sensor data from a primary sensor, such as a LIDAR or image sensor, and second sensor data from additional sensors, which may include LIDAR, image, GPS, IMU, or rotary encoder sensors. The system fuses these data streams to generate fused sensor data, which provides a more accurate localization of the system within its environment. Using this localization, the system accesses map data that describes objects in the environment, each object being associated with properties like position, shape, or type. The system then compares the first sensor data against these object properties to detect anomalies, such as discrepancies between observed sensor readings and expected values from the map. This comparison helps identify sensor malfunctions, environmental changes, or other irregularities, improving the reliability of autonomous navigation and decision-making. The fusion of multiple sensor modalities enhances robustness, while the anomaly detection mechanism ensures that the system can detect and respond to unexpected conditions.
18. The one or more non-transitory computer readable media of claim 17 , wherein determining the calibration parameter comprises: determining the calibration parameter based at least in part on a generative probabilistic model and using the map data, the map data further comprising information indicative of past measurements of the object made using the plurality of additional sensors.
This invention relates to systems for calibrating sensors in autonomous vehicles or robotic systems using probabilistic models and historical sensor data. The technology addresses the challenge of accurately calibrating sensors to ensure reliable perception and navigation, particularly when environmental conditions or sensor performance vary over time. The system involves a computer-implemented method that processes sensor data from multiple sensors to determine calibration parameters for a primary sensor. The calibration process leverages a generative probabilistic model, which statistically estimates the relationships between sensor measurements and environmental features. The model incorporates map data, including historical measurements of objects detected by additional sensors, to refine calibration accuracy. By analyzing past sensor readings stored in the map data, the system improves calibration robustness against noise, sensor drift, or environmental changes. The probabilistic model may use techniques such as Bayesian inference or machine learning to adjust calibration parameters dynamically. The map data includes metadata about object locations, sensor positions, and measurement uncertainties, enabling the system to correlate current sensor readings with past observations. This approach enhances sensor fusion and reduces calibration errors, improving the reliability of autonomous navigation and object detection. The invention is particularly useful in applications requiring high-precision sensor alignment, such as self-driving cars, drones, or industrial robotics.
19. The one or more non-transitory computer readable media of claim 15 , wherein the second sensor data is acquired from the LIDAR sensor, the first sensor data and the second sensor data corresponding to an object in an environment, wherein the first sensor data comprises a first distance or a first reflectivity, the second sensor data comprises a second distance or a second reflectivity, and determining that the first sensor data is anomalous comprises determining a difference between the first sensor data and the second sensor data.
This invention relates to systems for detecting anomalies in sensor data, particularly in autonomous vehicle or robotic navigation systems. The problem addressed is the need to accurately identify and handle discrepancies between sensor measurements, which can arise from environmental factors, sensor noise, or object properties. The system uses multiple sensors, including a LIDAR sensor, to collect data about objects in an environment. The first sensor data, such as distance or reflectivity measurements, is compared to the second sensor data, also comprising distance or reflectivity measurements from the LIDAR sensor. Anomalies are detected by calculating the difference between these measurements. If the difference exceeds a threshold, the first sensor data is flagged as anomalous. This allows the system to filter out unreliable data, improving navigation accuracy and safety. The method ensures that sensor discrepancies are identified before they affect decision-making processes, such as obstacle avoidance or path planning. The invention is particularly useful in autonomous vehicles, drones, or robotic systems where precise environmental perception is critical.
20. The one or more non-transitory computer readable media of claim 15 , the operations further comprising: receiving additional sensor data associated with a second measurement made by the calibrated first sensor, the second measurement being associated with an environment; generating, based at least in part on the additional sensor data, a trajectory, the trajectory configured to cause an autonomous vehicle to traverse a portion of the environment; and transmitting the trajectory to the autonomous vehicle.
This invention relates to autonomous vehicle navigation systems that utilize calibrated sensor data to generate and transmit trajectories for vehicle movement. The technology addresses the challenge of ensuring accurate and reliable sensor measurements in dynamic environments, which is critical for safe and efficient autonomous vehicle operation. The system involves a method for processing sensor data from a first sensor that has been calibrated using reference data from a second sensor. The calibration process adjusts the first sensor's measurements to improve accuracy, compensating for environmental factors or sensor drift. Once calibrated, the first sensor collects additional data related to a specific environment, such as road conditions, obstacles, or other relevant features. This data is then used to generate a trajectory—a planned path or route—that guides an autonomous vehicle through the environment. The trajectory is designed to optimize factors like safety, efficiency, and adherence to navigation constraints. After generation, the trajectory is transmitted to the autonomous vehicle, enabling it to follow the planned path autonomously. The invention ensures that the autonomous vehicle operates with high precision by leveraging calibrated sensor inputs, reducing errors in navigation and decision-making. This approach enhances reliability in real-world conditions where sensor accuracy is crucial for safe autonomous driving.
Unknown
November 10, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.